Long-term clinical and radiographic outcomes of anatomic total shoulder arthroplasty with a short-stem humeral component: a ten-year follow-up study

PURPOSE: This study aimed to provide long-term clinical and radiographic follow-up data for anatomic total shoulder arthroplasty using a short-stem humeral component with a cemented keeled glenoid in primary glenohumeral osteoarthritis. METHODS: Between 2010 and 2013, 35 consecutive anatomic total shoulder arthroplasties were performed at a single institution by one surgeon. At a mean follow-up of 10.3 ± 1.1 years, seven patients had died and five were lost, leaving 23 shoulders (82% of surviving cases) available for independent clinical and radiographic analysis. Clinical evaluation included the Constant Score, age- and sex-adjusted Constant Score, Subjective Shoulder Value, pain, and range of motion. Standardized radiographs at four time points were assessed for humeral remodeling, glenoid radiolucency, and secondary rotator cuff insufficiency. RESULTS: Clinical outcomes improved significantly compared with baseline and remained stable through long-term follow-up. The mean Constant Score improved from 29.5 ± 5.6 preoperatively to 72.4 ± 7.4 at final follow-up, with 16 shoulders (69.6%) graded as excellent. Radiographically, moderate-to-severe humeral remodeling was present in 11 shoulders (47.8%) at final follow-up, but remodeling was not associated with inferior clinical outcomes. Secondary rotator cuff insufficiency, assessed indirectly by acromiohumeral distance, was present in seven shoulders (30.4%). Higher glenoid radiolucency was associated with reduced forward elevation and abduction. One patient (4.3%) underwent revision 13.2 years after the index procedure due to aseptic glenoid loosening with secondary cuff insufficiency and was converted to reverse total shoulder arthroplasty. CONCLUSION: Anatomic total shoulder arthroplasty with a cemented keeled glenoid and a bone-preserving short-stem humeral component provided durable pain relief and functional improvement beyond ten years in this cohort. Humeral remodeling was frequent but clinically well tolerated, whereas glenoid radiolucency appeared to be associated with inferior long-term functional parameters and represented the only cause for revision in this cohort.